The Biostatistics Resource is a joint effort between Roswell Park Cancer Institute and the Department of Biostatistics at the University at Buffalo. The objective of the Biostatistics Resource is to be strategically involved in the design, implementation, and analysis of scientific data generated by CCSG program members. This includes the articulation of study objectives, statement of hypotheses, the derivation of appropriate designs and models for the cost effective achievement of a study's objectives, the assurance that modern and appropriate statistical methods are used, monitoring interim and final analyses, and assisting in the writing of abstracts and manuscripts. The expertise provided by the staff ensures that the yield of useful information from the scientific studies conducted at Roswell Park Cancer Institute is maximized while the costs are minimized. With Alan Hutson, PhD, Chair of Biostatistics at both the University at Buffalo and Roswell Park Cancer Institute, Gregory Wilding, PhD, Assistant Chair, and several newly appointed PhD and MS level biostatisticians, the Resource has improved dramatically with respect to its technical expertise. This expertise spans biostatistics, molecular epidemiology, bioinformatics and statistical genetics. Most Resource projects require the use of standard and commercially available tools. More recently, with the explosion of high throughput technologies, particular problems require an innovative solution for which the Resource develops novel methodology and/or custom software. Within the Biostatistics Resource, the Biotechnical, Analytic and Visualization Resource (BAVR), headed by Daniel Gaile, PhD, has been developed to spearhead software development for complex statistical genetics and bioinformatics problems. The Resource's expanded and enhanced capabilities are made known to CCSG Program leaders through referrals, by the Scientific Review Committee's requirement that protocols and grant applications have input from a biostatistician, the mandate by funding agencies that proposals include statistical designs with proper power analyses, formal seminars given by Resource staff, and presentations at Program meetings. Strong synergies have also developed in terms of the relationship with the Statistics and Data Coordinating center for the Gynecologic Oncology Group (GOG). In addition, the Resource has been a fertile ground for training graduate students in biostatistics. The Resource is used by all six Programs. $251,993 in CCSG support is requested, representing 22% of the total operating budget.